How parthbhatia1395/customer-segmentation-project works
• Data Source: Represented by the "data.csv" file. This is your raw invoice-level dataset.
• Data Ingestion Module: A component where data is read into the system (often using Pandas in the notebook).
• Data Preprocessing & Feature Engineering: This module is where the raw data is transformed into user-level features like total quantity, total spend, visit frequency, recency indicators, etc.
• Analysis and Visualization: Though not explicitly separated in file structure, include an output/evaluation module where the clusters are analyzed, interpreted, and potentially visualized.
• Notebook Environment: The entire workflow is executed within a Jupyter Notebook (Customer_segmentation_Parth_Bhatia_complete.ipynb), which can be shown as an orchestration layer that ties the previous components together.
Relationships and data flows
• Optionally, display the Notebook Environment encompassing all these steps, highlighting that this workflow is managed interactively.
Architectural patterns and design principles
• Pipeline/Flow-based design: Emphasize the sequential data transformations—from raw data to feature engineering and then to model training and evaluation.
• Clear separation of concerns within the notebook: Though all parts may exist in one file, conceptually they represent distinct stages.
Technologies
• Python (implied by Jupyter Notebook) and libraries such as Pandas (for data processing) and scikit-learn (for KMeans clustering).
• You might also label potential visualization libraries (e.g., matplotlib, seaborn) if they are used to present the segmentation results.